Excellent Presentation on Headphone Measurements

[Dan Clark]

Well I was surprised to see this, but cool!

There are tons of topics that can be covered, but as an intro I wanted to make an effort to make it accessible to non-engineers.

No, I won't spill beans on other products, I try to avoid direct comments on competitors, it's generally bad form, especially if it's not positive.

We actually do a lot of work with impulse response, in fact when Amir measured Stealth I asked him to share that measurement because I thought it was worth noting.

 

[nate]

At 11:22 he mentioned that there are 4 headphones currently in the Market that come close to the full bandwith of the Harman Response. 2 are made by his company and the other 2 are made by Harman. I'm curious of the Harman products because the stealth is quite costly..

 

[ADU]

At 11:22 he mentioned that there are 4 headphones currently in the Market that come close to the full bandwith of the Harman Response. 2 are made by his company and the other 2 are made by Harman. I'm curious of the Harman products because the stealth is quite costly..

(With apologies to Dan) my guess would be the AKG K371 and maybe the Mark Levinson 5909. Both of which are labels owned by Harman. Harman has released a number of headphones which are tuned to their targets under their various labels though.

The K371 and 5909 are two that have received a fair amount of attention though of late.

 

[Miiksuli]

Wow, I just learned something. Did not know that 150-250 Hz distortion is that u kinda wand if u want more slam on the bass. Interesting.

 

[Dan Clark]

Wow, I just learned something. Did not know that 150-250 Hz distortion is that u kinda wand if u want more slam on the bass. Interesting.

It's easy to test with PEQ. Also, if you go over 300hz you introduce "mud" and crush soundstage.

 

[Robbo99999]

Yeah, nice presentation. The slam colouration on the Stealth was interesting when he related it to frequency response (hump from 80-150Hz) and how that decision was also influenced by the very low distortion of the Stealth. It was also interesting to hear him mention a rough boundary for audible distortion in headphones, I think he mentioned 0.5%, although I'm assuming he's referring to distortion at the most sensitive hearing frequencies. I liked his explanation on the removal of standing waves by the Stealth design which he explained enabled a smoother less jagged response in the high frequencies and in terms of how that relates to the experience of listening to things like violins and how you want that high frequency stuff to be smooth to hear violins with more accuracy - that resonates with me when comparing listening to Max Ricthter's Four Seasons Recomposed on headphones (EQ'd to Harman through Oratory measurements) vs my anechoically flat EQ'd JBL 308p speakers, I do enjoy the violins more on those speakers than when listening on headphones...although I came up with this realisation maybe about a year ago when I was using my K702's more, so not checked my experience with other headphones I own. And this smooth high frequency he was talking about the Stealth, if you look at Amir's measurements I'd say it is less jagged than most planars I've seen measured, planars seem normally particularly bad in this area, but the Stealth does look smoother in the 1kHz-10kHz area, however not necessarily smoother in that region than a good dynamic driver headphone, Stealth frequency response follows:

 

[Koo]

My guess would be the AKG K371 and maybe the Mark Levinson 5909. Both of which are labels owned by Harman. Harman has released a number of headphones which are tuned to their targets under their various labels though.

The K371 and 5909 are two that have received a fair amount of attention though of late.

Can't be the K371, he said the Stealth is the only one that doesn't use DSP. My guess would be the JBL Tune 710 BT. https://www.audiosciencereview.com/forum/index.php?threads/new-jbl-closed-back-measurements.23843/

 

[Miiksuli]

It's easy to test with PEQ. Also, if you go over 300hz you introduce "mud" and crush soundstage.

I agree. This is why I do only lower frequencies to increase bass.

 

[odyo]

Excellent.

 

[odyo]

I don't know who you are referring to but I already show headphone distortion in percentage:

And it is shown at three different levels so you can see the progression. I don't know what you mean by distance though. These are headphones so distance is constant.

Can percentage somehow increase(worsen) at lower volumes ?
Mask more at higher volumes or behave better at higher volumes maybe ?

 

[Robbo99999]

What do you think are the other 2 headphones of the 4 he mentioned made by Harman that follow the curve? Could they be the k371 and k361?

I remember having the K371 spark into my mind when he said that. K371 has a big dip at 4kHz though, as being the only real place where it deviates from Harman. Maybe also the JBL Tune 710:

 

[odyo]

Yeah, nice presentation. The slam colouration on the Stealth was interesting when he related it to frequency response (hump from 80-150Hz) and how that decision was also influenced by the very low distortion of the Stealth. It was also interesting to hear him mention a rough boundary for audible distortion in headphones, I think he mentioned 0.5%, although I'm assuming he's referring to distortion at the most sensitive hearing frequencies. I liked his explanation on the removal of standing waves by the Stealth design which he explained enabled a smoother less jagged response in the high frequencies and in terms of how that relates to the experience of listening to things like violins and how you want that high frequency stuff to be smooth to hear violins with more accuracy - that resonates with me when comparing listening to Max Ricthter's Four Seasons Recomposed on headphones (EQ'd to Harman through Oratory measurements) vs my anechoically flat EQ'd JBL 308p speakers, I do enjoy the violins more on those speakers than when listening on headphones...although I came up with this realisation maybe about a year ago when I was using my K702's more, so not checked my experience with other headphones I own. And this smooth high frequency he was talking about the Stealth, if you look at Amir's measurements I'd say it is less jagged than most planars I've seen measured, planars seem normally particularly bad in this area, but the Stealth does look smoother in the 1kHz-10kHz area, however not necessarily smoother in that region than a good dynamic driver headphone, Stealth frequency response follows:

I find this ''bass punch'' hump very smart. I always try to do this when i eq my headphones otherwise the punch can be drown out. I generally tone down the 60-80hz area. That part makes too much constant, fatiguing sound.
So little less around 60-80, little more around 90-120 compared to harman.

 

[Robbo99999]

I find this ''bass punch'' hump very smart. I always try to do this when i eq my headphones otherwise the punch can be drown out. I generally tone down the 60-80hz area. That part makes too much constant, fatiguing sound.
So little less around 60-80, little more around 90-120 compared to harman.

I might experiment with mimicking that 80-150Hz hump in some of my headphones that have been measured. He did intimate that the main reason for including that hump was due to the fact that the Stealth is a very low distortion headphone, so it doesn't display the harmonics in that area that would happen as a result of bass distortion happening from tones lower down the frequency range in headphones that have higher distortion - as in the theory that distortion in the low bass of some headphones reinforces overall bass level by distortion creating harmonics in the bass further up the frequency range. So the Stealth wouldn't be able to get bass reinforcement by these distortion harmonics in the bass & therefore they added in the extra bass to make up for that. It could be the Harman Curve was created on headphones whereby the bass is partially reinforced by the harmonics that come about from the distortion that might have been present in those headphones during the process of creation of that curve. (Although wouldn't the amplitude of the harmonics be captured in the overall frequency response measurements of the headphone anyway, which would go against the reasoning behind the inclusion of that hump, even though Dan did say that was why they included the hump). But anyway, I might experiment incorporating that hump into some of my EQ's to see how it sounds......perhaps I'd expect it to fair the best in headphones that have low bass distortion, as that was supposedly the driving factor behind inclusion of this hump......I'd probably try it on my HE4XX (planar, and quite low distortion in bass), and also on my HD560s (dynamic driver, but low distortion in bass too).....I might try it on my closed back NAD HP50 (also low distortion) but it's hard to get the seal right on the HP50, so a bit unreliable to test on that one.

 

[SRKRAM]

Really enlightening talk, thanks Dan. Very interesting and well explained how harmonic distortion can give the impression of detail. Explains why a particular brand whose headphones have very high distortion claim that they are the most detailed.

 

[ClicketEKlack]

Really enlightening talk, thanks Dan. Very interesting and well explained how harmonic distortion can give the impression of detail. Explains why a particular brand whose headphones have very high distortion claim that they are the most detailed.

This may explain the reviews of a lot of high-end DACs as well -- those that measure poorly, but get raves for detail retrieval. R2R, etc.

 

[Phoney]

Nice presentation! But what about leakage effetcs? He briefly mentions it a couple of times but he doesn't really talk about it much. Harman studies has shown that some headphones could lose a lot of bass due to leakage if you can't get the optimal seal (which could matter a lot, even without things like huge beards or glasses involved). This seems like something worth talking more about when we consider how tonality is percieved to the user vs. the measured frequency response itself. Studies has shown that even some open back headphones can suffer to some extent due to sealing issues, but for closed backs the effect can be very noticeable. For those who use IEMs, you have probably experienced this effect to a high degree when using a pair of tips that were too small or didn't give you the best sealing.

 

[192kbps]

Which two did he already mention, nate?

 

[sai]

Kudos to Dan for the excellent, enlightening talk!

While FR is certainly most important, we are slowing learning more and more about how other measurements can give us insights about the more nuanced aspects of sound, as Dan indicated, those like soundstage, detail, etc. Maybe these terms just lack the definite measurement index as frequency response does tonality, but as the knowldge in this field proliferates I am sure we'll know better.

Personally I look at 'audiophile terms' like soundstage, slam and detail as synthetic descriptive terms that could involve one or more of the indices like FR, phase, THD, IR, GD etc. As someone who does measurements I feel it's sometimes not because these things are not measurable, but just that our knowledge and ability to interpret the graphs still need improvement.

 

[ADU]

Perhaps this deserves its own thread, but this was another interesting talk from the same conference given by the designer of the Sennheiser HE-1 and HD600/650/800, Axel Grell, who left Sennheiser awhile ago, and started his own company, Grell Audio.

It looks like one of the products he is developing is an equalization program that will allow people to more easily dial in the best sound for their own ears. The questions in the 2nd half were wide-ranging though, and covered alot of different topics, mostly related to his work at Sennheiser.

I thought he had some pretty interesting things to say on the subject of acoustic impedance as well, which is something I don't know nearly as much about as FR. He seemed to imply that measurements of this characteristic could potentially be up their with FR in terms of importance in the design and tuning of headphones though... Some very interesting stuff.

 

[ADU]

I gave Dan's video a 2nd watch as well, and think I came away even more disappointed than on the first listen. I'm glad if some others here were able to take something useful away from it though. Different strokes as they say.

I have alot of respect for Dan's skills as a headphone designer and developer though.